Compare / Muse Wonder vs GPT-6 Luna
Muse Wonder vs GPT-6 Luna
Which is cheaper, which is more capable, and which does better at each kind of work, from 65 results on 4 sources that measured both models.
MetaMuse Wonder
OpenAIGPT-6 Luna
Muse Wonder wins 1820 too close to callGPT-6 Luna wins 27
Of the 65 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.
Results as of 13 October 2026 (catalogue release 2026-10-13-2e2420e3c5cc). See also Meta vs OpenAI, every model from both labs.
Muse Wonder or GPT-6 Luna? The short answer
Pick Muse Wonder if quality matters most: it scores higher (Spring Prompt overall 70 against 40). Pick GPT-6 Luna if cost matters more: it costs 76 times less per product listing on CatalogBench.
- Spring Prompt overall: Muse Wonder 70, GPT-6 Luna 40 (out of 100; how the score works).
- GPT-6 Luna is cheaper to run: $0.000 against $0.036 per product listing on CatalogBench (76 times less).
- Muse Wonder is newer: released 12 Oct 2026, against 22 Sep 2026 for GPT-6 Luna.
- On our own benchmarks, on the results both models have: Muse Wonder does better for product listings; GPT-6 Luna for marketing planning.
Head to head
70
Spring Prompt overall
40
$8.00
Price per million tokens, blendedlower is better
$0.20
1,048,576
Context window, tokens
1,050,000
Every fact side by side
| Muse Wonder | GPT-6 Luna | |
|---|---|---|
| Price per million tokens | $4.00 in · $20.00 out | $0.10 in · $0.50 out |
| Blended (3 in : 1 out) | $8.00 | $0.20 |
| Cheapest host, blended | – | $0.20 (Azure) |
| Cost per product listing (CatalogBench) | $0.036 | $0.000 |
| Cached input, per million | $0.20 cached input | $0.010 cached input |
| Intelligence Index | – | 38.1 |
| Spring Prompt overall | 70 | 40 |
| Output speed, tokens a second | – | 127 |
| Context window | 1,048,576 tokens | 1,050,000 tokens |
| Released | 12 Oct 2026 (newer) | 22 Sep 2026 |
| Weights | Closed | Closed |
| Developer based in | the US | the US |
Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.
Which is better for what
Each model's rank among the models on sale today, for each kind of work, from our own benchmarks and licensed sources. Further right is better; the better rank is in green.
Product listingsTurning a sparse product feed and photos into listings that can go live
4 of 21
5 of 21
Decks from an analysisTurning a finished analysis into a deck you could present as it is
8 of 20
6 of 20
User surveysPlanning a user survey and reading its results without being misled
2 of 9
not measured
Marketing planningPlanning a year of ad spend without overspending
10 of 20
3 of 20
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls
not measured
43 of 164
Professional workReal tasks from banking, consulting and law, business documents and freelance projects
not measured
25 of 51
Reasoning and knowledgeHard questions across science, maths and general knowledge
not measured
40 of 198
WritingWhat people prefer in blind comparisons, and judged writing quality
not measured
60 of 188
Sticking to the factsSummarising without inventing things, and factual answers
not measured
94 of 152
Speed under pressureGood decisions against a real clock (fast chess)
12 of 24
19 of 24
Their slides, side by side
The same brief for both: Manchester store investment review for Hearthside Coffee, an invented company. First slide of each deck from DeckBench.
From our research
- Claude Opus 5.5 vs Muse Wonder Full: Eight Crafts, and Opus Still Wins Most 13 Oct 2026
- Muse Wonder review: good analysis, work that still needs checking 13 Oct 2026
- AI Checkpoint: Sol’s new fast lane 9 Oct 2026
- We Raced AI Models to Find One Small Thing in a Huge Picture. Cutting It Up More Than Doubled What They Found. 8 Oct 2026
Every shared result
Each model's best result on every measure both have, grouped by benchmark, with the setting that produced it. Under each measure, every model that benchmark has measured is a grey tick, best to the right, so you can see where the two sit in the field. The headline measures of each benchmark come first; the rest are under "Show all". Green marks the better value; where a source reports ranges that overlap, neither is marked.
| Measure | Muse Wonder | GPT-6 Luna |
|---|---|---|
| BulletBench measured by Spring Prompt · 3 results | ||
| Cost per game · Blitz 3+2US dollars, lower is better 12 · 1 of 14 measured | $0.27 | $0.0072 |
| Cost per game · Bullet 60sUS dollars, lower is better 28 · 5 of 29 measured | $0.0663minimal reasoning | $0.0016no reasoning |
| Ladder Elo · Bullet 60sladder Elo, higher is better 19 · 24 of 29 measured | 409minimal reasoning | 249no reasoning |
| CatalogBench measured by Spring Prompt · 3 results | ||
| Cost per productUS dollars, lower is better 17 · 1 of 21 measured | $0.0361 | $0.0005 |
| Reliably publish-ready% of products, higher is better 3 · 6 of 21 measured | 66.1% | 55.4% |
| Reliably publish-ready · sales brief% of products, higher is better 4 · 4 of 21 measured | 51.8% | 51.8% |
| DeckBench measured by Spring Prompt · 3 results | ||
| Cost per deckUS dollars, lower is better 14 · 1 of 20 measured | $0.18 | $0.0087 |
| Deck ratingrating, higher is betterThe rating weighs each win by how strong the other deck was; the win rate counts wins against whichever decks a model happened to meet, so the two can disagree. 8 · 6 of 20 measured | 1,273 | 1,288 |
| Accurate decks% of tasks, higher is better 1 · 9 of 20 measured | 100.0% | 50.0% |
| ROASBench measured by Spring Prompt · 3 results | ||
| Overall scorescore out of 100, higher is better 10 · 3 of 20 measured | 38.9 | 53.4 |
| Audience scorescore out of 100, higher is better 12 · 3 of 20 measured | 54.7 | 68.3 |
| Business scorescore out of 100, higher is better 10 · 3 of 20 measured | 34.5 | 54.8 |
Show all 65 results (53 more)
| Measure | Muse Wonder | GPT-6 Luna |
|---|---|---|
| BulletBench measured by Spring Prompt · 7 results | ||
| Games lost on time · Blitz 3+2% of games, lower is better 14 · 11 of 14 measured | 75.0% | 50.0% |
| Games lost on time · Bullet 60s% of games, lower is better 23 · 19 of 29 measured | 58.3%minimal reasoning | 50.0%no reasoning |
| Invalid moves · Blitz 3+2% of moves, lower is better 1 · 1 of 14 measured | 0.0% | 0.0% |
| Invalid moves · Bullet 60s% of moves, lower is better 1 · 1 of 29 measured | 0.0%minimal reasoning | 0.0%no reasoning |
| Ladder Elo · Blitz 3+2ladder Elo, higher is better 13 · 6 of 14 measured | 358 | 557 |
| Median move time · Blitz 3+2milliseconds, lower is better 12 · 11 of 14 measured | 7.1 s | 7.0 s |
| Median move time · Bullet 60smilliseconds, lower is better 25 · 19 of 29 measured | 1.9 sminimal reasoning | 1.5 sno reasoning |
| CatalogBench measured by Spring Prompt · 27 results | ||
| Channel compliance% of products, higher is better 14 · 1 of 21 measured | 99.4% | 100.0% |
| Channel rules broken% of products, lower is better 11 · 1 of 21 measured | 0.6% | 0.0% |
| Channel rules broken · sales brief% of products, lower is better 1 · 1 of 21 measured | 0.0% | 0.0% |
| Claims to check% of products, lower is better 1 · 1 of 21 measured | 0.0% | 0.0% |
| Claims to check · sales brief% of products, lower is better 1 · 1 of 21 measured | 0.0% | 0.0% |
| Conflicts caught% of conflicts, higher is better 1 · 1 of 21 measured | 100.0% | 100.0% |
| Content quality% of checks, higher is better 3 · 10 of 21 measured | 98.3% | 97.0% |
| Content quality · sales brief% of checks, higher is better 3 · 9 of 21 measured | 98.5% | 97.4% |
| Decision accuracy% of decisions, higher is better 3 · 8 of 21 measured | 98.3% | 97.0% |
| Failed outputs% of products, lower is better 1 · 1 of 21 measured | 0.0% | 0.0% |
| Failed outputs · sales brief% of products, lower is better 1 · 1 of 21 measured | 0.0% | 0.0% |
| Field accuracy% of missing fields, higher is better 1 · 9 of 21 measured | 96.1% | 93.7% |
| Invented values% of filled values, lower is better 11 · 13 of 21 measured | 1.4% | 1.7% |
| Missing UK information% of products, lower is better 1 · 1 of 21 measured | 0.0% | 0.0% |
| Missing UK information · sales brief% of products, lower is better 1 · 1 of 21 measured | 0.0% | 0.0% |
| Not findable% of products, lower is better 4 · 18 of 21 measured | 3.6% | 10.1% |
| Not findable · sales brief% of products, lower is better 8 · 18 of 21 measured | 3.6% | 10.1% |
| Publish-ready listings% of products, higher is better 3 · 7 of 21 measured | 75.0% | 67.9% |
| Publish-ready listings · sales brief% of products, higher is better 5 · 4 of 21 measured | 67.3% | 67.9% |
| Unsupported claimsclaims per product, lower is better 4 · 5 of 21 measured | 0.21 | 0.31 |
| Unsupported claims% of products, lower is better 4 · 5 of 21 measured | 7.1% | 8.9% |
| Unsupported claims · sales briefclaims per product, lower is better 5 · 4 of 21 measured | 0.36 | 0.35 |
| Unsupported claims · sales brief% of products, lower is better 5 · 4 of 21 measured | 17.9% | 8.3% |
| Wrong attributes% of products, lower is better 2 · 12 of 21 measured | 13.1% | 18.5% |
| Wrong attributes · sales brief% of products, lower is better 4 · 12 of 21 measured | 14.3% | 19.1% |
| Wrong category or variant% of products, lower is better 10 · 4 of 21 measured | 2.4% | 1.8% |
| Wrong category or variant · sales brief% of products, lower is better 10 · 1 of 21 measured | 2.4% | 0.0% |
| DeckBench measured by Spring Prompt · 13 results | ||
| Caveat dropped% of tasks, lower is better 1 · 1 of 20 measured | 0.0% | 0.0% |
| Clean layout% of tasks, higher is better 18 · 6 of 20 measured | 16.7% | 66.7% |
| Design quality% of the maximum, higher is better 9 · 5 of 20 measured | 67.1% | 78.1% |
| Draft figure quoted% of tasks, lower is better 1 · 12 of 20 measured | 0.0% | 33.3% |
| Findings missing% of tasks, lower is better 1 · 1 of 20 measured | 0.0% | 0.0% |
| Head-to-head win rate% of comparisons, higher is betterThe rating weighs each win by how strong the other deck was; the win rate counts wins against whichever decks a model happened to meet, so the two can disagree. 10 · 9 of 20 measured | 63.2% | 65.2% |
| Layout defects% of tasks, lower is better 18 · 6 of 20 measured | 83.3% | 33.3% |
| Misleading metric used% of tasks, lower is better 1 · 1 of 20 measured | 0.0% | 0.0% |
| Presentable decks% of tasks, higher is better 10 · 12 of 20 measured | 16.7% | 0.0% |
| Recommendation late or wrong% of tasks, lower is better 1 · 8 of 20 measured | 0.0% | 16.7% |
| Slides needing work% of tasks, lower is better 15 · 2 of 20 measured | 66.7% | 16.7% |
| Unsupported claims% of tasks, lower is better 1 · 1 of 20 measured | 0.0% | 0.0% |
| Unsupported numbers% of tasks, lower is better 1 · 1 of 20 measured | 0.0% | 0.0% |
| ROASBench measured by Spring Prompt · 6 results | ||
| Consistency scorescore out of 100, higher is better 12 · 4 of 20 measured | 25.9 | 35.8 |
| Contribution profitsimulated US dollars, higher is better 11 · 4 of 20 measured | $489,176 | $738,879 |
| Cost of a runUS dollars, lower is better 15 · 1 of 20 measured | $0.57 | $0.0142 |
| Months over budgetmonths of 12, lower is better 1 · 1 of 20 measured | 0 | 0 |
| Planning scorescore out of 100, higher is better 4 · 11 of 20 measured | 55.0 | 54.8 |
| Return on ad spendprofit per $1 spent, higher is better 11 · 4 of 20 measured | $0.41 | $0.66 |
Compare them with other models
- Meta vs OpenAIEvery model from both labs, best against best
- Muse Wonder vs Claude Opus 5.5Highest-scoring overall at $3–10 per million tokens
- Muse Wonder vs GPT-6 AstraJust above overall
- Muse Wonder vs GPT-6 SolJust below overall
- Muse Wonder vs Muse Spark 1.3Meta's best other model
- GPT-6 Luna vs Claude Haiku 5.5Most intelligent at under $0.25 per million tokens
- GPT-6 Luna vs DeepSeek-V4.1-FlashFastest with similar intelligence
- GPT-6 Luna vs Claude Opus 5.5Most intelligent available today
Quick answers
Is Muse Wonder better than GPT-6 Luna?
Pick Muse Wonder if quality matters most: it scores higher (Spring Prompt overall 70 against 40). Pick GPT-6 Luna if cost matters more: it costs 76 times less per product listing on CatalogBench.
Which is cheaper, Muse Wonder or GPT-6 Luna?
GPT-6 Luna is cheaper to run: $0.000 against $0.036 per product listing on CatalogBench (76 times less).
Which is better for product listings, Muse Wonder or GPT-6 Luna?
Muse Wonder ranks 4 of 21 models on sale for turning a sparse product feed and photos into listings that can go live, against 5 for GPT-6 Luna.
Which is better for decks from an analysis, Muse Wonder or GPT-6 Luna?
GPT-6 Luna ranks 6 of 20 models on sale for turning a finished analysis into a deck you could present as it is, against 8 for Muse Wonder.
Which is better for marketing planning, Muse Wonder or GPT-6 Luna?
GPT-6 Luna ranks 3 of 20 models on sale for planning a year of ad spend without overspending, against 10 for Muse Wonder.
Which is better for speed under pressure, Muse Wonder or GPT-6 Luna?
Muse Wonder ranks 12 of 24 models on sale for good decisions against a real clock (fast chess), against 19 for GPT-6 Luna.
Which has the bigger context window, Muse Wonder or GPT-6 Luna?
GPT-6 Luna: 1,050,000 tokens, against 1,048,576 for Muse Wonder.
Run Muse Wonder and GPT-6 Luna on your own prompt
Benchmarks aren't your data. Try both side by side in the playground, with the cost of every answer.